1 |
On Wed, Aug 24, 2011 at 07:30:11AM +0000, Robin H. Johnson wrote: |
2 |
> Bandwidth: Along the same lines, rsync will always be able to use less |
3 |
> bandwidth than Git, because none of the intermediate commits need to be |
4 |
> transfered. This will be esp. evident as a user tree gets older (the |
5 |
> amount of mtime/checksum metadata scales linearly with the size of the |
6 |
> tree, not the age of the tree. |
7 |
|
8 |
Yes, rsync protocol scales with the project size. |
9 |
|
10 |
> The actual file content transfered scales linearly with the age of the tree). |
11 |
|
12 |
(assuming you are talking about git here) |
13 |
Uhm no? git protocol scales with chageset size. You don't retransfer |
14 |
already transferred content. So you will be tranferring only content |
15 |
from the last git pull. |
16 |
|
17 |
rsync checks the whole tree everytime. The extra intermediate commits |
18 |
that git sends out might be a lot smaller than cheking the whole tree |
19 |
(and will be if you git pull frequently). I find the "always" in the |
20 |
first line highly surprising. Did you make any tests? |
21 |
|
22 |
-- |
23 |
Eray Aslan <eras@g.o> |